Quantum machine learning, secure quantum computing and other applications using integrated quantum photonics

Abstract

This talk presents recent experimental demonstrations that use integrated nanophotonic processors for various quantum computations such as quantum machine learning and in particular reinforcement learning, where agents interact with environments by exchanging signals via a communication channel. We show that this exchange allows boosting the learning of the agent. Another experiment underlines the feasibility of such photonic integrated processors for a homomorphically-encrypted quantum walk computation. This secure quantum computation exploits path- and polarization as degrees-of-freedom for encrypting the input and output of the photonic processor. As last demonstration I will present counter-intuitive quantum communication tasks that are linked to the Zeno effect. As outlook I will discuss technological challenges for the scale up of photonic quantum computers, and our group’s current work for addressing some of those.